Table 1 Performance of synthetic data
From: Model-based learning of information diffusion in social media networks
Network type | Model | Nodes | Edges | Cascades | Train accuracy | Test accuracy |
|---|---|---|---|---|---|---|
Erdos-Renyi | LT | 50 | 335 | 200 | 76.4% | 65.6% |
| Â | RW | 50 | 335 | 200 | 98.9% | 68.8% |
| Â | LTNN | 50 | 335 | 200 | 72.0% | 69.6% |
| Â | RWNN | 50 | 335 | 200 | 76.6% | 70.7% |
Barabasi-Albert | LT | 50 | 525 | 200 | 77.6% | 63.1% |
| Â | RW | 50 | 525 | 200 | 99.5% | 68.3% |
| Â | LTNN | 50 | 525 | 200 | 75.8% | 71.8% |
| Â | RWNN | 50 | 525 | 200 | 79.1% | 73.1% |
Watts-Strogatz | LT | 50 | 350 | 200 | 76.6% | 62.9% |
| Â | RW | 50 | 350 | 200 | 98.6% | 67.5% |
| Â | LTNN | 50 | 350 | 200 | 72.6% | 70.8% |
| Â | RWNN | 50 | 350 | 200 | 76.6% | 68.9% |